Meshes & Interference
5:40 PM -- I’m going to wrap up my current work on WiFi and interference with this one final blog entry and its accompanying Tech Note, which you can find here. This time, though, I don’t have any hard data; I just want to pose a couple of questions. The topic is the deployment of metro-scale WiFi meshes and the potential interference that might be caused to them by residential and enterprise (and hot-spot) systems already installed, and, potentially much more importantly, from meshes to these other systems.
We’ve already seen that two WLAN systems operating in close proximity can indeed cause mutual degradation, and the effect here can be dramatic no matter whether the load is general network traffic, VoFi, or VidFi. [Ed note: Fi D'oh!] So it stands to reason that some interference can be expected in both directions. My theory at the moment is that interference from external WiFi traffic (of any form) to the mesh is unlikely to have much of an impact. Mesh nodes, the vendors tell me, make use of sophisticated RF spectrum management techniques, and can work around any interference they see. Resilience, after all, is one hallmark of a wireless mesh of any form, so it stands to reason that we should be all set here. I do worry about interference from the opposite direction, however -- meshes operate at much higher transmit power levels than typical residential and enterprise APs, and the use of multiple radios per node means that the other guys could be in deep trouble should a mesh node be installed nearby.
I think the 5GHz bands will provide some refuge in the near term. But, longer term, all WLANs will need to integrate spectrum assurance (SA) capabilities into their management suite, analyzing interference of all forms and automating the resolution of problems when they appear. As meshes proliferate, the challenge only gets worse, at least in urban areas. By the way, all of this led to another thought -- just how do we evaluate the performance of large-scale meshes, anyway? That’s a problem I’m just starting on now; more on this later this year.
— Craig Mathias is Principal Analyst at the Farpoint Group , an advisory firm specializing in wireless communications and mobile computing. Special to Unstrung
We’ve already seen that two WLAN systems operating in close proximity can indeed cause mutual degradation, and the effect here can be dramatic no matter whether the load is general network traffic, VoFi, or VidFi. [Ed note: Fi D'oh!] So it stands to reason that some interference can be expected in both directions. My theory at the moment is that interference from external WiFi traffic (of any form) to the mesh is unlikely to have much of an impact. Mesh nodes, the vendors tell me, make use of sophisticated RF spectrum management techniques, and can work around any interference they see. Resilience, after all, is one hallmark of a wireless mesh of any form, so it stands to reason that we should be all set here. I do worry about interference from the opposite direction, however -- meshes operate at much higher transmit power levels than typical residential and enterprise APs, and the use of multiple radios per node means that the other guys could be in deep trouble should a mesh node be installed nearby.
I think the 5GHz bands will provide some refuge in the near term. But, longer term, all WLANs will need to integrate spectrum assurance (SA) capabilities into their management suite, analyzing interference of all forms and automating the resolution of problems when they appear. As meshes proliferate, the challenge only gets worse, at least in urban areas. By the way, all of this led to another thought -- just how do we evaluate the performance of large-scale meshes, anyway? That’s a problem I’m just starting on now; more on this later this year.
— Craig Mathias is Principal Analyst at the Farpoint Group , an advisory firm specializing in wireless communications and mobile computing. Special to Unstrung
